Herasight Develops Genetic Tests for Embryos That Can Reduce Disease and Boost IQ

In an exclusive conversation ahead of SynBioBeta 2026, Jonathan Anomaly frames the discussion around a simple reality: parents are already making choices that influence their children’s genes when they select a mate. In many cases, those decisions are becoming more intentional, informed by the science of polygenic prediction.

When it first emerged in 1978, IVF was considered controversial. The advent of genetic tests for embryos in the 1990s sparked a fresh bout of controversy. But controversy yielded to acceptance 

Herasight decided to push the boundaries even farther in 2023 when they founded their company in order to develop predictors for complex diseases, as well as traits like intelligence. They expected fierce pushback but, when Herasight emerged from stealth in 2025,  “the anticipated outrage didn’t occur,” Anomaly says. “Culture had moved on.” 

Anomaly is not a conventional biotech founder. After earning a doctorate in philosophy and teaching courses at universities like Duke and Penn, he has spent years working through the social and ethical implications of reproductive technologies like those his company is now offering. Some of his recent writing focuses on how heritable traits like intelligence and empathy influence pro-social behavior. 

Herasight translates that perspective into a product. The company builds polygenic risk scores that allow women who are doing IVF to minimize the risk of a variety of diseases; but they can also influence non-disease traits like intelligence. 

The challenge of navigating all this information is real. Parents have to weigh probabilities across many traits at once. As Anomaly puts it, the difficulty lies in “taking complex and probabilistic information into account when selecting an embryo to implant.” Herasight also features a tool that aggregates information about each disease into a longevity index across embryos. 

Rather than validating their scores using unrelated individuals from large biobanks, the company tests whether its scores can correctly predict the traits of adult siblings. “If polygenic scores can predict the traits of adult siblings,” Anomaly says, “then we know they work in the context of IVF, where embryos share the same percentage of DNA with each other as well as the same parents.”

The company also emphasizes integration across traits. Some worry that selecting one trait – like a lower risk of diabetes – might increase the risk of some other undesirable trait. But according to a recent publication, Herasight showed that “antagonistic pleiotropy” is rare. Instead, positive pleiotropy is common, meaning that selecting against one disease often lowers the risks of other diseases as well. One example is that “selecting an embryo with a low risk of schizophrenia also tends to lower the risk of severe depression and bipolar disorder.” 

Embryo selection has been possible for decades through IVF and preimplantation genetic testing. What has changed is the ability to model complex traits. Polygenic scores, built from large-scale genome-wide association studies, estimate risk across many variants. These predictions remain probabilistic and do not capture all environmental effects, but their accuracy continues to improve as datasets expand.

In the near term, selection is likely to drive progress while gene editing remains limited. “Multiplex editing… may be quite a few years away,” Anomaly says. For the next phase of reproductive genomics, embryo selection is expected to carry most of the practical impact. This aligns with current constraints: selecting among embryos fits within the existing IVF process, while genome editing at scale still faces serious technical barriers.

Herasight is also interested in a key bottleneck. The effectiveness of selection depends on the number of embryos available. Each IVF cycle produces a limited set, which restricts potential gains from polygenic screening. Even accurate models usually yield modest improvements when the choice set is small.

This is why Anomaly highlights in vitro gametogenesis (IVG) as a priority. IVG would enable the generation of eggs or sperm from somatic cells, expanding the number of embryos available for selection. A larger pool increases genetic variation and improves selection outcomes.

These trends position embryo selection as a key feature for future reproductive medicine.

Herasight’s core claim is that predictive models can inform decisions at the point of embryo selection. Its success will depend on model performance and on whether parents choose to use this information in practice.

At SynBioBeta 2026, Anomaly will expand on this vision, focusing on the longevity index and the challenge of making probabilistic genomics usable in high-stakes decisions. His talk will explore what is possible today and what comes next as reproductive technologies scale.